-
717
ISBN: 978-93-80689-28-9
Effect of Process Parameters on Weld Bead Geometry and Element
Segregation in
Electron Beam Welding of INCONEL 718
Manu Thomas[a]*, Partha Saha[b], Girish Namboothiri[a],
Radhakrishnan G[a], Snehil Srivastava[a]
[a]Vikram Sarabhai Space Centre, Trivandrum
[b]IIT Kharagpur, West Bengal, India
___________________________________________________________________________________
Abstract
INCONEL alloy 718 is a widely used super alloy in aerospace
industry for applications involving high temperatures up to
650°C.
Welding of INCONEL 718 is heavily depends on Electron Beam
Welding (EBW) as key joining technique. Thus extensive
characterization of EBW of Inconel 718 is required. In this
present work, we study the effect of various process parameters
viz. beam
current, beam focus, welding speed etc. on the quality of welded
coupons based on the observations from etched bead geometry.
The
larger underfill that found in sharp-focus samples compared to
the up-focus samples is explained using Marangoni effect. At
higher
welding speeds uniform penetration and minimum underfill are
observed. The extent of alloy element segregation due to changes in
weld
parameters and the effect of process parameters on weld bead
geometry are investigated. We also discuss the reasons for changes
in bead
geometry and their implications.
Images from a Scanning Electron Microscope (SEM) reveal the
presence of Nb rich phases in Austenitic matrix of weld bead.
Formation
of Nb rich phases in weld region is detrimental as it causes a
drop in the percentage of Niobium (Nb) in bulk matrix. Nb
segregation in
root region was found to be less as compared to that of top
region of the weld bead. Tensile strength of specimens were
determined and
these were in concurrence with the SEM observations on element
segregation. Superior tensile strength values are observed for
sharp
focus cases as compared to up-focus cases. This behavior is
further investigated by correlating changes in laves formation with
the beam
focus changes.
Keywords: Electron Beam Welding (EBW), Inconel Alloy 718, Heat
Affected Zones (HAZ)
______________________________________________________________________________________________________________
1. INTRODUCTION
Inconel 718 is a Ni–Cr–Fe-based super alloy. It is widely
used
in aerospace industry for applications that involves high
temperatures up to 640°C. It can be easily fabricated, and
possess good tensile, creep, fatigue, rupture strength. It is
used
in manufacture of fasteners and instrumentation parts. It has
got
good resistance to post weld cracking which is an important
welding characteristic that increases its applicability.
Inconel alloy 718 shall be welded by Gas tungsten arc
welding
(GTAW), Laser beam welding (LBW) and Electron Beam
Welding (EBW). Although the EB welding has some inherent
advantages over other welding techniques on Inconel-718, the
mechanical properties of EB welds are still considerably
lower
compared to the parent material (PM) properties. However,
studies indicated that significant advantages in terms of
mechanical properties can be achieved by controlling Laves
phase formation and altering its morphology by employing
beam oscillations techniques in EB welding. The mechanical
properties of welded components are influenced by the
microstructure of fusion zone (FZ), and heat-affected zone
(HAZ).
Electron Beam Welding (EBW) is a fusion welding in which
joint is made by heating the work piece due to impingement
of
the focused electron beam of very high kinetic energy on the
work piece. When the electron beam hits the work piece,
kinetic
energy of the electron beam is getting converted into
thermal
energy, resulting in melting and even evaporation of the
work
material.
Janaki Ram et al. [1] observed that, the strength of Inconel
718
mainly depends on the gamma prime phase (γ’- phase)
precipitation that occurs during heat treatment. Inconel 718
utilizes Niobium (Nb) as the prime strengthening element.
The
addition of Nb avoids the problem of strain age cracking and
increases the weld strength. On the other hand, the separation
of
Nb in the inter dendrite areas, that occurs during the
solidification process associated with the welding creates a
severe constraint. This separation of Nb leads to the
formation
of an inter metallic phase called “Laves phase” represented
as
(Ni, Fe, Cr)2 (Nb, Ti, Mo, Si)3. The more the Nb separates
out
in the inter-dendrite regions, the larger is the volume of
laves
phase that is formed (Vincent, 1985) [2].
E.G.Vinayan et al [3] studied the effect of electron beam
welding of solution treated Inconel 718 evaluating its
tensile
properties, fracture toughness, weld bead microstructure and
micro hardness. Further to this, responses of the aforesaid
parameters in EB weld specimens are compared with parent
material properties.
GAO Peng et al. (2011) [4] observed that after solution and
two-aging treatment of electron beam welded Inconel 718
super
alloy thick plate, the γ-phase (Ni3Nb) precipitates in the
grain
boundary. Once the precipitation of δ-phase (equilibrium
orthorhombic Ni3Nb precipitate) increases, γ' (Ni3(Al,Ti))
and
γ'' (Ni3Nb) will decrease because Nb is shared by all of
them
(Saied Azadian et al., 2004)[5]. The stress rupture and
tensile
properties of the weld are inferior compared to the coupons
prepared with solution pre-treatment when aging heat
treatment
is done on weld coupons without post weld solution heat
treatment, (Janakiram et al., 2004)[1].
Saied Azadian et al. (2004) observed that the maximum rate
of
δ phase precipitation happens at temperatures close to 900°C
and have confirmed full dissolution of δ phase beyond the
solvus temperature range. Improved mechanical properties are
assured if the laves creation is controlled by a appropriate
method during EB welding. As an example, a beam oscillation
method can be used to decrease laves phase formation.
Madhusudhana Reddy et al. (2009) [6] found that with
appropriate beam oscillation, the laves morphology can be
positively changed to get improved weld properties.
-
718
J.K. Hong et al. [7] studied the microstructures and
mechanical
properties of Inconel 718 welds by CO2 laser welding. They
found that for Inconel 718 plate having 5mm thickness,
minimum 6 kW of laser power was required to produce full
penetration welds without defects such as porosities. This
was
taken as the base for setting the range of heat input required
to
get full penetration for our 6.6mm sample. The paper also
mentions that the optimum focus position for maximum weld
depth and defect-free weld was on the surface (sharp focus
beam)
2. EXPERIMENTAL SETUP The EBW machine used for welding the
samples is
EO Paton make KL134 model. The governing parameters of
this particular machine are as follows:
• Chamber pressure: 2.6x10-4 torr
• Distance between EB gun and top surface of specimen: 245mm
• Focusing current: o sharp focus: 547mA, up focus by 2mm:
555mA & up focus by 4mm:563mA
• Power: 30 kW
• Accelerating voltage: 60 kV Inconel alloy 718 plates of 6.5 mm
thickness were
selected for the study. 36 plates of size 75x 80x 6.5 mm
were
taken with suitable edge preparation and cleaning.
A theoretical heat input value was calculated based on the
following assumptions:-
• Since thermal conductivity of Inconel 718 is very less
(11.4 W/Km), heat loss due to conduction can be
neglected.
• Since experiment was carried out in vacuum of the order 2 to 3
x 10-4torr, heat loss due to convection is not
present.
• Heat loss due to radiation was also neglected.
• All material properties were assumed to remain same during
welding.
• For welding to take place, temperature should be more than
liquidus temperature and less than boiling
temperature. Thus the weld pool was assumed to be at a
mean temperature of this.
Thus it was assumed that all heat supplied was used to
melt the material i.e. to form the weld pool.
The following equation was used:- 𝜂𝑃
𝑣= 𝐴𝜌(𝑐𝑝Δ𝑇
∗ + 𝐿𝑓)
Where η = overall efficiency
P = power
v = welding speed
A = weld cross sectional area
ρ = density of material
Cp = specific heat capacity at constant pressure
ΔT* = mean temperature of liquidus and boiling
temperatures i.e. (Tl + Tb)/2
Lf = latent heat of fusion
The expected keyhole weld profile was plotted
graphically with width of weld assumed as 4mm and with full
penetration of 6.5mm.From this weld area calculated was 12.5
mm2. Also overall efficiency was assumed as 0.9 considering
the fact that heat loss associated with EBW of low
conducting
material like Inconel 718 is very minimum. Substituting
these
values in the above equation, theoretical heat input was
found
as 𝑃
𝑣= 121.76 𝐽/𝑚𝑚
We have Power, P = V*I
Where V = accelerating voltage = 60 kV &
I = beam current in mA
The experiment was carried out for 3 cases: -
Case 1. Sharp focus
Case 2. 2mm upfocus
Case 3. 4mm upfocus
For each case, 5 welding speeds were selected as
15mm/s, 20mm/s, 25mm/s, 30mm/s and 32mm/s and the
corresponding beam currents were calculated.
Case 1: Sharp focus
At welding speed of 25mm/s and corresponding beam
current of 50.733 mA, a trial run was carried out with bead
on
weld to validate these parameters. It was found that full
penetration was not achieved with these parameters. Hence
beam current was increased to 58 mA and subsequently to 62
mA, 66 mA and 70 mA with welding speed remaining the
same. It was found that the best weld (good penetration and
less
underfill) was achieved with 66 mA beam current. In case of
actual weld, beam current required is slightly higher than
the
case for bead on welds. Hence a beam current of 68 mA was
selected for Case 1: Sharp focus at the welding speed of
25mm/s. Correspondingly actual heat input required also
changed to 163.2 J/mm. Using this heat input, beam currents
for
other welding speeds were calculated (as in table 1) and
welding was carried out.
Table 1 Welding Parameters for Case 1
Set 1 Set 2 Set 3 Set 4 Set 5
Scan
speed(mm/s)
15 20 25 30 35
Beam
current(mA)
40.8 54.4 68 81.6 95.2
Case 2: 2mm upfocus
A bead on weld trial run was carried out at welding speed
of 25mm/s and beam current same as that in case 1 i.e. 68
mA.
The weld was found to be good (good penetration and fewer
underfills). Hence the same parameters as used in Case 1
were
used in this case also and good welds were obtained for
every
welding speed.
Case 3: 4mm upfocus
Fig 1. Samples after cutting to required dimensions
-
719
Again a bead on trial weld was carried out at welding speed
of
25mm/s and beam current of 68 mA. However it was found that
the weld was very poor. There was a drastic decrease in
depth
of penetration. Hence beam current was increased to 72 mA
and
subsequently to 76 mA, 80 mA, 84 mA, 88 mA, 92 mA, 96 mA
and 100 mA with welding speed remaining same at 25 mm/s. It
was found that the best weld (good penetration and less
underfill) was achieved with 100 mA beam current. For actual
welding, this was increase to 102 mA. Corresponding actual
heat input required also changed to 244.8 J/mm. With this
heat
input value, beam currents for other welding speeds were
calculated (as in table 2) and welding carried out.
Table 2 Welding Parameters for Case 3
Set 1 Set 2 Set 3 Set 4 Set 5
Scan
speed(mm/s)
15 20 25 30 35
Beam
current(mA)
61.2 81.6 102 122.4 142.8
3. RESULTS AND DISCUSSIONS
3.1 Effect of Process Parameters on Weld Bead Geometry
The specimens after chemical etching were analyzed
under a microscope to assess weld profile characteristics
viz.
width of weld, underfill, weld depth (as shown in Figure2)
etc.
This was done to understand the effect of various process
parameters on weld bead geometry.
Figure 2 shows the weld profile attributes
Figure 3 Bead width Vs speed graph for different focus
The bead width of weldment depends on the beam focus
conditions. Samples prepared with up-focused beam have a
broader reinforcement at the top compared to the samples
done
with a sharp-focused beam. Nevertheless, for a specific
focus,
the welding speed did not make a significant impact on weld
bead width (Figure 3)
Reinforcement/under fill and root side penetration are
plotted
against speed as shown in figure 4.a & 4.b.With the
welding
parameters which give partial weld depth, reinforcement is
observed in the bead geometry. From the Fig.4 a & b it can
be
observed that at 15 mm/s weld speed, reinforcement is
present
for both up focus cases (2mm and 4mm up focus) where the
depth of weld is less than plate thickness. As the welding
speed
increased to 20mm/s we get non uniform penetration and high
underfill in the bead geometry. But when the speed increases
beyond 20mm/s, underfill is reduced and resulted in uniform
penetration. Thus from these observations it is evident that
at
welding speed 30-35mm/s bead geometry is more uniform with
less underfill. The similar trend was observed for sharp
focus,
2mm up focus and for 4mm up focus.
(a)
(b)
Figure 4 (a) Reinforcement/underfill Vs speed graph (b)
Penetration Vs speed graph for different focuses.
For the electron beam heat source, a Gaussian power density
distribution is approximated with the heat intensity being
largest at the centre and falling towards the periphery [5]. As
a
result, a differential temperature exists in the molten weld
pool
with highest temperatures at the centre. This leads to the
surface
tension gradient within the weld pool. Molten metal at the
weld
centre that has the lowest surface tension is pulled in the
direction of the regions with higher surface tension
(Marangoni
effect). This causes a depression in the molten weld pool
and
leads to the formation of weld underfill. In the case of
sharp
focused beam, a larger surface tension gradient exists inside
the
molten weld pool compared to the samples prepared by an up-
focus beam. This explains the underfill profile that found
in
sharp-focus samples compared to up-focus samples (Figure 5a
& 5b).
The area of cross section of weld bead is computed for all
the
samples and theoretical heat input was calculated. The
cross-
section area of the bead is increasing linearly with increase
in
beam current, thus theoretical heat input calculated based
on
bead area is also increasing. From figure 6a & 6b it can be
seen
that the theoretical heat input line is always at an offset from
the
actual heat input line. This may be due to the various heat
losses due to conduction, vaporisation and the assumptions
used
while calculating theoretical heat input.
-
720
(a) (b)
Figure 5(a): Underfill in sharp focus sample
Figure 5(b): Underfill in 4mm up focused sample
Figure 6 a Heat input Vs Beam current (sharp focus)
Figure 6 b Heat input Vs Beam current (4 mm up focus)
It is found that for sharp focus beam the difference between
actual and theoretical heat input increases with beam
current.
This may be due to the increased vaporisation and other
losses
at higher beam current, since the heat intensity is also
increases
with beam current at sharp focus.
Figure 7 shows how theoretical heat input is related to the
welding speed. It can be seen that as the welding speed
increases, the theoretical heat input computed based on the
cross-section area of the weld bead increases slightly. It may
be
due to the less conduction/evaporation losses at higher
speeds
compared to the lower speeds. The difference between the
actual heat input and theoretical heat input attributes to all
the
heat losses including conduction losses, vaporization losses
and
the assumptions made during theoretical heat input
calculation.
Figure 7 Heat input Vs speed graph
In the case of 4 mm up focused (figure 8), beam
current/heat input required for full penetration was much
more
than that of sharp focus. This may be due to the reduced
heat
intensity at 4 mm up focus condition. In order to get full
penetration through key hole formation some critical heat
intensity is required below which depth of beam penetration
falls drastically. At 4 mm up focus, heat intensity might
have
fallen below that critical value since the beam area at the
surface is more than that of sharp focused beam. In order to
increase the heat intensity at the joint, the beam current has
to
be increased, which results in an increased heat input to
the
weld bead.
Figure 8 Heat input vs weld speed graph for 4mm up focus
3.2 Weld Pool Microstructure Observations Using SEM
In austenitic matrix of weld bead, the presence of Nb rich
phases is revealed in SEM images (Figure 9). This phase
contains significantly more Nb than the parent material. The
formation of Nb rich phase in the weld region is harmful as
it
causes a reduction in the percentage of Nb in bulk weld
matrix,
which is a vital strengthening element in the matrix.
Fig 9: SEM image of weldment
40
60
80
100
120
140
160
180
20 25 30 35
hea
t in
pu
t (J
/mm
)
Speed (mm/s)
Heat input(theoretical)Sharp focus
Heat input(theoretical)2mm up focus)
Heatinput(actual)
Nb rich phase
-
721
Weight percentage of various alloy elements at the bulk
matrix of the top and root portions of the weld samples are
computed from an Energy Dispersive Spectroscopy (EDS)
analysis. The observations are taken for both sharp and
upfocus
samples separately and the results are summarized in Table
3.
Due to the presence of Nb rich phases, Nb percentage in weld
matrix (~3.6%) is small compared to the parent metal
(~5.6%).
Table 3: Weight percentage of various element in top and bottom
region
Parent Metal
Sharp Focus,
25mm/s,Top
Sharp Focus,
25mm/s,Root
4mm Up focus,
25mm/s,Top
4mm Up focus,
25mm/s,Root
Element
Weight% Weight% Weight% Weight% Weight%
Ti K 1.12 0.78 0.98 1.05 0.83
Cr K 19.27 19.36 18.96 19.35 18.91
Fe K 18.77 20.51 19.96 20.13 19.46
Ni K 51.47 53.06 51.97 52.88 52.77
Nb L 5.66 3.14 3.61 2.73 4.31
Mo L 3.71 3.14 4.52 3.86 3.72
The percentage of Nb in Nb rich phase at top region of 4 mm
up-focused samples is more compared to that of root region
(21.86 % and 16.89 % respectively) (Figure 10 & 11).
This
leads to more Nb deficiency in the top region of samples
welded with up-focused beam. At the root region Nb
deficiency
is less than the top region. Higher Nb content in Nb rich
phases
at the top portion of weld makes it more brittle and incapable
of
handling thermal stresses during welding [9].
Element Weight%
Ti K 2.07
Cr K 15.35
Fe K 13.64
Ni K 47.09
Nb L 21.86
Fig 10: Nb rich phases at top region
Element Weight%
Ti K 1.32
Cr K 16.53
Fe K 14.81
Ni K 46.35
Nb L 16.89
Fig 11: Nb rich phases at root region
3.3 Tensile Testing Results
In this stud, a subsize specimen was used for tensile
testing.
Accordingly, specimen dimensions were set as per ASTM
E8/E8M-13: "Standard Test Methods for Tension Testing of
Metallic Materials" (2013). From each weld coupons, 3
tensile
specimens were made and tensile strength of specimens were
found out using universal testing machine. The results are
as
given in figure-12. For the given heat input, the difference
in
tensile strength of samples is found to be marginal.
For those specimens welded with a sharp focused beam,
the tensile strength is marginally high (~892 MPa) compared
to
the up focused weld specimens (~872 Mpa). This variation in
tensile strength can be due to the difference in Nb
separation
associated with laves phase formation which affects the weld
-
722
strength. In sharp focus samples Nb separation is less
compared
to samples welded with up focused beam.
Fig 12: UTS Vs Welding Speed
4 CONCLUSION
The effect of process parameters in electron beam
welding of Inconel 718 plates are investigated based on
bead geometry observations. SEM/EDS analysis revealed the
presence of Nb rich phases in Nb deficient matrix. More
Nb segregation is observed at top portion of the weld bead
compared to the root region. Tensile testing measurement is
also taken and the results were confirmatory with the
SEM/EDS
findings. Following are the major conclusions made.
1) The underfill is observed in all the weld specimens welded at
various process parameters. Nevertheless, the underfill
is more in samples that are welded with sharp focused
beam. This is described using Marangoni effect of
temperature on surface tension of liquids.
2) The difference observed between actual heat input and
theoretical heat input may be due to the various heat losses
(conduction, radiation, vaporization) and the various
assumptions made during the theoretical calculation.
3) Bead width remains the same for all beam current and welding
speed combinations for a particular heat input and
it increases with increase in heat input.
4) Reinforcement/underfill is affected by penetration. At high
penetration, high underfill is observed and for insufficient
penetration, reinforcement was observed.
5) The formation of Nb rich phases during the solidification of
weld pool causes a drop in the percentage of Nb in the
weld matrix compared to the parent metal. The Nb
separation is more in the upper region of weld bead which
is welded with up-focused beam compared to the weld
samples welded at sharp focus.
6) The presence of brittle laves phases and higher Nb
segregation from the bulk matrix leads to the tensile
strength of up-focus samples being smaller than that of the
sharp-focus samples.
References
[1] Janaki Ram, G.D., Venugopal Reddy, A., Prasad Rao, K.,
Madhusudhan Reddy, G., Microstructure
and mechanical properties of Inconel 718 electron
beam welds, Mater. Sci. Techno. 21, 1132–1138
(2005).
[2] Vincent, R., Precipltation around welds in the Nickel-base
superalloy, inconel 718. H. H. Wills
Physics Laboratory, University of Bristol (1985).
[3] E.G Vinayan,Mahesh Babu. B. R,Satheeshkumar A,Vinu Paul, R
Sarath, Shibu Gopinath.,Electron Beam welding of
Inconel 718,VSSC-MS 11-268. [4] GAO Peng, ZHANG Kai-feng, ZHANG
Bing-gang,
JIANG Shao-song, ZHANG Bao-wei,
Microstructures and high temperature mechanical
properties of electron beam welded Inconel 718
superalloy thick plates, Trans. Non Ferrous Met.Soc.
China 21, s315-s325 (2011).
[5] Saied Azadian, Liu-Ying Wei, Richard Warren., 2004. Delta
phase precipitation in Inconel 718.,
Materials Characterization 53, 7– 16.
[6] Madhusudhana Reddy, G., Srinivasa Murthy., C.V., Srinivasa
Rao, K., Prasad Rao, K., Improvement of
mechanical properties of Inconel 718 electron beam
welds-influence of welding techniques and postweld
heat treatment. Int J Adv Manuf Technol 43:671–
680 (2009).
[7] Hong, J.K., Park, J.H, Park, N.K., Eom, I.S., Kim, M.B.,
Kang, C.Y., Microstructures and mechanical
properties of Inconel 718 welds by CO2 laser
welding, Journal of materials processing technology.
201,515–520 (2008).
[8] Paolo Ferroa, Andrea Zambonb, Franco Bonolloa.,
Investigation of electron-beam welding in wrought
Inconel706—experimental and numerical analysis.,
Materials Science and Engineering A 392, 94–105
(2005).
[9] Richard, N.L., Hung, X., Chaturvedi, M.C., Heat affected
zone cracking in cast Inconel 718. Mater.
Charact. 28, 179–187 (1992).
[10] Huanga, C.A., Wang, T.H., Lee, C.H., Han, W.C., a study of
the heat-affected zone (HAZ) of an Inconel
718sheet welded with electron-beam welding
(EBW). Materials Science and Engineering A 398,
275–281 (2005).
[11] Radhakrishna, C.H., Prasad Rao, K., Srivivas, S., Laves
phase in superalloy 718 weld metals. J Mater
Sci Lett 14:1810–1812 (1995).
[12] Thompson, R.G., Mayo, D.E., Radhakrishnam, B., Relationship
between carbon content, microstructure
and intergranular liquation cracking in cast nickel
alloy 718. Metall. Trans. A 22, 557–567 (1991).
[13] Vishwakarma, K.R., Richards, N.L., Chaturvedi, M.C., HAZ
microfissuring in EB welded ALLVAC
718 plus alloy. In: Procedings of Superalloy 718,
625 and Derivatives, pp. 637–647 (2005).
[14] Girish.N.Namboodiri., Manu Thomas., Abhinand Karna., Snehil
Srivastava., Paul G Panicker., Shibu
Gopinath., quality assessment of electron beam welds of
inconel-718 plates using phased array
ultrasonics and its comparison with radiography
inspection (2015)
860
870
880
890
900
20mm/s 25mm/s 30mm/sec
Ten
sile
Str
engt
h (
MP
a)
Welding speed (mm/s)
4mm upfocus
Sharp focus